近红外光谱技术在中药栀子渗漉液理化指标快速分析中的应用研究

Rapid analysis of Zhizi percolation liquid mixture physical and chemical indicators with NIR spectroscopy

用声光可调滤光器(AOTF)-近红外(NIR)光谱法在线分析栀子渗漉液密度及栀子苷含量。在线收集栀子渗漉液样品,建立栀子渗漉液的含量、比重数据库,同时采集近红外光谱图谱,用偏最小二乘(PLS1)法分别建立NIR光谱与含量、比重数据之间的校正模型,并对在线过程中收集的预测集样品进行含量预测来验证所建模型。结果表明,渗漉液NIR光谱与含量、比重数据之间的校正模型相关系数R2分别为0.956 0和0.978 2,外部样品预测平均相对偏差分别为0.09%,2.84%;该方法精密度RSD分别为0.06%,2.96%;稳定性RSD分别为0.63%,1.89%;预测回收率分别为100.0%,99.9%。近红外光谱技术在栀子渗漉液密度及淫羊藿苷含量分析中具有快速、直接、多成分同时测定,并能实现现场分析。

Specific density and geniposide content of Zhizi（fructus gardeniae） percolation fluid were analyzed with acousto-optic-tunable-filter（AOTF）-near infrared（NIR） spectroscopy technology.Zhizi percolation fluid samples were collected online.Database of specific density and geniposide content of Zhizi percolation was established,including NIR spectroscopy.Calibration model of NIR spectroscopy-specific density,and NIR spectroscopy-geniposide content were established with partial least squares（PLS 1）,the calibration model prediction with samples collected online.The proportion of the data correlation between the calibration model coefficients R^2 were 0.956 0 and 0.978 2;the average relative deviation of prediction outside the sample were 0.09%,2.84%;method precision RSD was 0.06%,2.96%;stability of the RSD were 0.63%,1.89%;forecast recoveries were 100.0%,99.9%.NIR spectroscopy for online analysis of specific density and geniposide content of Zhizi is fast,direct,simultaneous determination of multiple components,and to achieve on-site analysis.